CA1133520A

CA1133520A - Production of methyl and ethylamines with rhodium- iron catalysts

Info

Publication number: CA1133520A
Application number: CA366,377A
Authority: CA
Inventors: William J. Bartley
Original assignee: Union Carbide Corp
Current assignee: Union Carbide Corp
Priority date: 1979-12-19
Filing date: 1980-12-09
Publication date: 1982-10-12
Also published as: EP0032623A3; EP0032623A2; DK540680A; JPS56120644A; NO803817L; US4250116A

Abstract

12,641 PRODUCTION OF METHYL AND ETHYLAMINES
WITH RHODIUM-IRON CATALYSTS

ABSTRACT OF THE DISCLOSURE
A heterogeneous process for electively producing monoamines in which the nitrogen is bonded only to methyl, ethyl, hydrogen or combinations thereof which comprises contacting a mixture of carbon monoxide, hydrogen and a nitrogen-containing compound selected from the group con-sisting of ammonia, nitric oxide and mixtures thereof with a heterogeneous solid catalyst comprising rhodium and iron.
The monoamines of the invention are formed in a collective amount of at least 50 weight percent of the total amine products of the reaction.

SPECIFICATION

1.

Description

3 ~
12,641 This invention relstes to a method of preparing monoamines ~n which the nitro~en is bonded only to methyl, ethyl, hydrogen or combinations thereof. More particularly, the invention relates to the prepara-tlon of such monoamines by the reaction of B 6ynthesis gas with ammonia and/or nitric oxide in the presence of a rhodium-iron catalyst.
The synthesis of aLky~ ines by the catalytic rea~tion of hydrogen with ~arious c~mbinations of reactants from among carbon monoxide or carbon dioxide, ammonia, nitrogen, aldehydes and primary amines is extensively described in the prior art. U.S. Patent :
3,597,438, for example, discloses the production of secondary amines by the reaction of hydrogen and an aliphatic aldehyde with ammonia or a primary amine in the presence of a rhodium metal catalyst; U.S~ Patent Nos. 3,636,153 and 3,646,148 are directed to processes for producing methylamine and d~methylamine by the reaction of hydrogen, carbon monoxite and nitrogen . .
over a zirconium or hafnium ca~alyst; and U.S. Patent Nos. 3,410,904 and 3,444,203 disclose the production of .
methylamines by the reaction of hydrogen, carbon monoxide and ammonia i~ the presence of catalysts such as copper, ~::
palladium, silver and plaeinum. The processes of these patents in common with most amine synthesis processes disclosed in the ar~ are general~y characterized by either the production of a wide spectrum of amine protucts. in-cluding many of relatively little commercial value, or ~ the synthesi.s of a very specific product mixture9 such as methylam~.nes. The present i~vention is directed to . ~ ,

2. :
~1 :

- -33 3~
12,641 the production of a commescl~lly desirable product mixture, namely, low m~lecular weight amines which ln-clude meehylamines B5 well as ethyl~m~nes and subs~itut-ed methyl and ~thyl un~nes. The present invPntion is predicated on the discovery that a caeal~st containing rhodium ant iron can 6electively produce such noamines to the substantial exclusion of higher carbon number amine products thereby aYoiting the need for extensi~e product purlficat$on.
Iron-containing catalysts for the reaction of synthesis gas with ammonia are known to produce a wide spectrum of alkyl amlne products. Thus, U.S. Patent ~o. 2,518,754 discloses the use of Fischer-TSopsch type catalyst such as ~etallic lron to produce a mixture of primary, secondary and tertiary A~ines containing up to 12 carbon atoms, with higher alkyl primary noamines predominating. Similarly, ehere is disclosed in U.S.
Patent No. 3,726,926, at c~lumn 6, Table 2, a very wide weight distribution of amine fractions which is typical for that fo~ned by the reaction of hydrogen, carbon monoxide and ammonia in the presence of a catalyst comprising iron oxide. Similar amine product distribu-tions formed with iron caealys~s are disclosed in the following publ~cations: Bashkiaov, et al.~ Doklaty Akad. Nauk. S.S.S.R.~ 109, 774-6 (1956), lChem. Abs., 51, 4931e (1975~]; Russian Patent 133,890, December 10, 1969 lChern. Abs., 55, 14308f (1961)~; and ~oelbel, et al., Angew, Chem. Internat. Ed., Vol. 5, 843 (1966), all of which publications disclose the psoduct~on of aLkyl mono~rnines ranging in chain length from 3 to . ~ .

3.

3~ ~'2~
~2,641 greater thsn 11 carbon atoms (Russlan Patemt 133,890) to as high a~ 40 carbon atoms tXoelbel, et ~1., 6upra.) Cstalysts contalning shod~um snd lron in com-bLnation are known, although not for the product~on of amines. For example, United Xlngclom Patent 1,501,B91, published February 22, 1978, discloses the u6e of Rh-Fe catalysts for the select~ve preparatlon of two-car~on atoms oxygenated compounds, partirularly ethanol~ from a synthesis gas mix~ure. A similar disclosure 18 found in an article by Bhasin, et alO~ Journal of Catalysis, Yol. 54, pp. 120-128 (1978). The use of shodium-iron c~talysts for ~he preparation of E~Lnes, has heretofore noe been contemplated. Indeed, Unitet Ringdom Patent 436,414 which describes a process for the production of tert~ary amines using hydrogenation catalysts from among Gsoups 1, 2, 6, 7 and 8 of the Periodic Table (iron being included in Group 8) specif~cally states at column 2, lines 85-87, that ~oble ~etals are not to be employed as catalysts for the particular seact~on therein disclosed.
.

SUMMARY OF THE INV~NTION
The process of the inveneion describes a catalyst for the ~elective production of monoamines in which the nitrogen is bonded only to methyl, ethyl, hydrogen or combinations thereof. The process involves contacting a heterogeneous solid cstalyst comprising rhodium and iron with a mixture comprising hydrogen, carbon monoxide and a r~itrogen-containing compound 6elected from the group cons$sting of ammonia, nitric oxide and mixt~res of came under suitable reaction condieions So

4.

.. ~

33 ~ '~
~2,641 produce the desired ~onoa~ines. The process of ~he invention ~s characterized by the formation of the above-descrlbed monoamines in a collective amount of at least 50 we. Z of the total amine products f~rmed.

- PR~CESS DISCUSSION
- The reactlo~ i~ conduce:ed at reactlon condi-tions of temperature, pressure, gas compos$tion and space ~eloci~y to produce the defined r~noam~nes in a collective amount which is at least about 50 wt. Z, desirably at 1~ least 75 wt. %, and under preferred reaction condit~ons at l~ast 90 wt. Z, of the total amine products formed by the reaction.
Conditions of temperature, of pressure, and of gas composition are usuaLly within the ranges that are essentially conventional for synthesis gas conversions such as those employed ~n the production of methanol.
Thus, existing technology and equipment may generally be used to effect the reac~ion.
The reaction is h$ghly exothermlc, with both the thermodynamic equilibrium ant the kinetic reaction rates being governed by the reac~ion t~mperature. . ;~`
Average catalyst ~ed t~mperatures are usually within the range of about 200-450C., but for optimum conversions, bed tempera~ures are kept within the range of about 250-400C., typically about ~50- 50C.
She reaction temperature is an lmportant process variable, affecting not only total productivity but selectivity towsrd the amine products. Over relatively narrow temperature ranges, as fos exsmple 10 or 20Co`~

~ ~ -:~1335Z~
~2,641 an increase in temperature mAy somewhat incre~se total ~ynthesis gAS conversion, tending to increase the production of amines. A~ the ~a~e time, however, higher temperatures favor methane production, which ~pparently $ncreases more rspidly ~t higher temperatures than does conversion of re~ctants to amine products. Thus, for a ~ glven catalyst and with all other variables held constant~
- the optimum temperature will depend ~ore on product and process economics thsn on thermotynamic or kinet~c consi-derations, with higher temperatures tending to increase the production of amine products but dispropor~ionatRly increasing the co-production of methane.
In the discuss~ons above, the indicated temper-atures are expressed as average, or mean9 reaction bed temperatures. Because of the exothermic nature of the reaction, it is desirable that the temperature be controlled 50 as not to produce a runaway methanation, in which methane formation is increased with higher temperature, and the result~ng exotherm increases ~he temperature further. To accomplish this 9 conventional -temperature control techniques are utilized, as for example the use of fluidized bed reaction zones, the use of multi-stage fixed bed adiaba~ic reactors with inter-stage cooling, or relatively small catalyst particles placed in tube-and-shell ~ype reactors with a coolant fluid surrounding the catalyst-filled tubes.
The reaction zone pressure is des~rably within the range of about 15 psig to about 10,000 psig, economically within the range of about 300-5,000 psig. In general, higher reaction zone pressures increase the total weight of 6.

- . . . . ... . .
~.

33 ~
12,641 product obt~ined per unit time and likewi6e ~mpr~ve the select$vity toward am~ne products.
The ratio of hydrogen to carbon monoxide in the synthesis gas may vary widely. Normally the mole ratio of hydrogen to carbon monoxide is within the r~nge of 20:1 to 1:20, preferably within the range of ~bout S:l to sbout 1:5. Increasing ehe ratio tends to ~ncrease the total rate of reaction 60metimes quite significantly, and has a smaller though favorable effect o~ the rate of production of ~m~ne products, but concurrently increases selectiYity to methane.
The percent conversion of C0 to products is an ~mportant process variable. At low conversions e.~., less than about one-fourth of the C0 per pass and preferably not more than about one-eighth, the formation of the amines of the present invention is increasingly favored relative to other products. This conversion is conveniently achie~ed by employing a h~gh space velocity correlated w~h other reaction Yariables (e.g., tempera-ture, pressure, gas composition and catalyst). Space ~elocities in excess of about 102 gas hourly space velocity (volumes of reactant gas, at 0C. ~nd 760 mm merc~ry pressure, per volume of cataiyst per ho~r, commonly referred to as "&HSV") ase generally employed, although i~ $s preferable that the space velocity be with~n th~ range of about 103 to about ~o6 per hour.
With regard to the amine products formed, increased space veloc~.ties favor the formation of primary and secondary amines while decreased space velocities, i.e., below 104 h~-. 1, favor the formation of tertiary amines.
~:
7.

..... .. .. . . ... ~.

. ~ ... .

3 ~
12,641 Tertiary amines be~ng the most ~table of the amlne protucts fDrmed will generally be pre~ent in the product mixture ~n ~ubstant~ally grealeer amD~ltS than primary and secondary amines.
The concentration of nitrogen-contalning compound in the reactant gas ~x~ure affects total productivity as well 8S the amine product diseribution. Generally, the effect of increaslng the concen~ration of æmmon~s and/or nitric oxide in the ~eed $s to decre~se the overall rate of reaction and increase the ~electivity of the reaction to amines. Conversely, lower concentrations of ammonia and/or ni~ric oxide e~hance the productlvity of the reaction but d~sproportionately favor the production of methane and alcohols thereby lowering the reaction selectivity to amines. With regard to the amine products formed, at a fixed conversion of CO, ~ncreasing the concentration of ammonia and/or n~tric oxide tends to favor the formation of primdry and secondary amines while decreasing the concentration of such nitrogen-containin~
compounds favors the formation of eertiary m;nes.
The Rh-Fe catalyst of the invention comprises rhodium ~n combination with iron upon a support material.
This is typically effected by depositi~g rhodium and iron onto a particulate support m2terial and plac~ng the supported c~mbination into the reaction zone. On the basis of experience to date, the amount of catalyst on the support should range from about 0.01 we~ght percent to sbout 25 weight percent, based on the combined weight ~ of the metals and the support material. Preferably, the amount of catalyst is within the range of from ~bout 0.1 ~o about~l0 ~eight percent.

L2,641 The weight ra~io of iron to rhodium in the catalyst should generally ~ary from about 0.01 to about 10 to produce alkyl amines in accordance with the inYention. The particular wei~ht ratio of iron to rhodium will affect the amine product distribution.
The use of relat~vely low weight ratios of iron to rhodium in the catalyst, for example~ a weight ratio of about 0.05 or less, results in the forma~ion of pre-dominantly ethylam~nes. Conversely, at a weight ratio of about 0.1 or 8reater the amine products are predo-minantly comprised Qf methylamines. The preferred weight ratio of iron/rhodium is within the range of from about 0.02 to ~bout 1.
A relatively high surface area particulate support, e.g., one having a surface area upwards of about 1.0 squ re meters per gram (BET low tempera~ure nitrogen adsorption isotherm method), is preferred, desirably up-wards of about 10 square meters per gram, although sur-face area alone ls not the sole determ~native variable.
Based on research to date, silica gel and titania are preferred as the ca~alyst base or support, with graphite, graphitized carbon, alpha alu~na, zirconia, magnesia, eta-alus~na, gamma-alumina, and active carbon being less desirable ~or the purpose of this invention, rhodium teposited on either particles of iron oxide or a carrier containing iron is substantially the same as rhodium and iron codeposited on any of the above support materia~s. ~`
The rhodium and iron may be deposited onto the catalyst base or support by any of the techniques .. . ... . ... . . .... . . .. . .... ... - .. .
- , " .

~ ~ 3 32~ 12,641 commonly used for catalyst preparation, as for example, impregnation from an organic or ~norgan~c 601uelon, pre-cipitation, coprec$pitation, or cation exchange. Con-veniently, a solution of a heat decomposable inorganic or organic rhodium compound and an iron co~pound ~s appropriately contacted w~th ehe support materlal, and the support then dried and heated, the latter advanta-geously under reducinR conditions, to form ehe finely dispersed iron-containing rhodium catalyst. Any of these materials may be deposited concurrently or sequentially.
The rhodium~ deposited is typically in metal form, desirably as fine discrete particles. The form of the iron component is, however, not completely appre-ciated. }t may be chemically associated wlth the rh~dium or it may be in a physical admixture with the rhodium.
For example, the iron may be alloyed with the shodium or not, in the form of a metal or an oxidized state of the metal, or it may be in the form of an oxide, a silicate, an a~uminate, a carbonate, or the like.

DESCRIPTION OF TEST REACTOR
The reactor used in these studies was a 316 `
8tainless steel, bottom-agitated '~agnedrive" autoclave ~ of the J.~l. Berty ~esign with a centrally positioned catalyst ~asket and a side product effluent line. It is of the type depicted in Figure 1 of the paper by Berty, Hambrick, M~lone and Ullock, entitled '~eactor for Vapos-Phase Catalytic Studies", presented as Preprint 42E at the Symposium on Advances in High-Pressure Technol~gy - ~

10. . .-33 ~20 12,641 Part lI, Sixty Fourth National Meeting of the Amerlcan Instltute of Chemical Engineers (AIChE), at ~ew Orleans, Loulsiana, on March 16-20, 1969, snd obtalnable from AIChE at 345 East 47th Street, New York~ N. Y.
10017. The autoclaYe was internally gold plated and the interior voLume was sbout 1 l~ter. A varla~le speed, magnetically driven fan continuously rec~rculated the reaction mixture over the ca~alyst bed. The following modifications were found to facilitate opera~ion and ~nhibit run-away methanation reactions 1. Hydrogen feed gas was introduced continu-ously at the bottom of the autoclave through the well for the shaft of the Magnedrive agitator.
2. Carbon mo~oxide feed gas was introduced continuously through a separate port at the bottom of the autoclave~ in order to avoid 8 hydroge~-rich zone ~n the autoclave.
Effluent gases were removed through 2 port in the side of the reac~or. Condensable l~quid products were remoYed from the exi~ 6tream in 8 brine-cooled condenser at ca. 5 to 10~. and were collected ~n a holding tank under pressure. After venting to atmos-pheric pressure, the non-condensable components were sampled through a rubber 6eptum for analysis. The exit seream was then sent through a,wet-test meter to determine its total volume. No external recycle ~as employed.

DESCRIPTION OF THE,TEST PROCEDURE
The bulk volume of the weighed catalyst sample was determined and the sample was placed in ~he ca~alyst 1~ 33 ~ 12,641 basket. The quantity of catE~lyst charged was 50 cc which provided an estimated reactant gas conversion of less than 10 percent. Gold-plated screens and thin layers of glass wool were placed above snd below the catalyst bed to prevent circulation of solid fines. The c~talyst basket was charged to the reactor, and the reactor then sealed. The sealed reactor and the process lines were pressure tested at ambient temperature and 1,000 psig using nitrogen, hydrogen, or a mixture of the two.
When the reactor was shown to be leak free, pure hydrogen was passed through the re~ctor at 1,000 psig and the temperature rais-ed to about 250C. The hydrogen and carbon monoxide flows were then adjusted at a mole ratio of 1:1 to give an approximate purge rate of 450 STP*
liters/hr. corresponding to a space velocity of about 9,000 STP volumes of gas per volume of catalyst per hour.
The H21CO ratio was determined by gas chrom~tographic analysis of an effluent gas aliquot.
When the desired gas composition ~as obtained, the reactor temperature was raised to 300DC. ~hen con- ~;
ditions had stabilized, concentrated ammonia was pumped (as a liquid) to the reactor at a rate of about 24 ml./hr. using an Altex Scientific Inc. Model 100 piston pump. A period of about one hvur was allowed for the reactor to reach a *STP refers to standard temperature and pressure defined at 0C. and 1 atmosphere pressure.

12.

33 ~2~
12,~41 ~teady 6tate at the new operatlng conditions before beginning to measure actu~l time of re~ction. A ~ample of llquid product was collected over a one-houx period by cooling the product-contalning gas ~n a brlne-chilled condenser at 1,000 psig snd ~hen collecting the liquid product in a Dne-liter stainless steel receiver. The liquid sample w~s then ~nalyzed ~y gas chIomatography, The non-condensable gases were metered through a wet-test meter to determine their ~olume9 snd a gas sample was collected and analyzed by gas chromatography to determlne its composition. The combined results are reported in Table I below.

CATALY~T PREPARATION
The cstalyst cited ln Table I below was prepared as follows:
Rhod~um trichloride and ferric chlor~de were dis-solved in one pore volume of distilled water at ambient temperature. Davison Grade 59 Silica Gel (8-20 mesh) was , placed in a vacuum flask. The top of the flask was sealed 2~ with a rubber septum, and the flask was evacuated through the side arm. A syringe needle was then used to inject the rhodium and iron solution onto the evacuated support while ~hsking the flask. When addition was complete, the impre~nated support was allowed to stand at one atmosphere for cs. 30 minutes~. It was then dr~ed in a nitrogen s~tmosphere as follows: 85C. tl hr.); 110C.
.~ (2 hrQ.); 150C. (2 hrs.~; 300C. (2 hr3.). The dried, impregnsted 3upport was placed in ~ quartz tube through which hydrogen was continuously passet. The temperature 13.

., . . ........... - - - -.
. ~

~3 ~2~
12,641 was raised to 500C. over a 3-hour perlod ~nt held at that value for 1 hour. The reduced cat~lyst was cooled to ~mbient temperature in an atmosphere of ~lowing hydrogen~
In order to remove signif~cant ~mounts of lmpurities which were present in the ~upport uaterial as - received from the ~anufacturer, the Davi80~ Grade 59 - silica support was initially '~ashed" with oxalic acid prior to being used as the catalyst support~ Such treat-ment consisted of passing a mixture of oxalic acid, glycerine, and wa~er ~n proportions of 1:1.5:2.5 by weight, respectively, through a bed of ~upport material (length/diameter ratio of about 20 to 25) contained within a glass tube which drained through a stopcock at its base. The contents of the tu~e were maintained at about 90C. by means of resistance heating wire wrapped around the exterior of ~he tu~e. About 2.5-volumes of oxalic acid 601ution were used to wash one volume of 8-20 mesh silica gel over a three-hour period. The material was then washed with about s~ volumes of distilled water at 90C. over a perlot of about four hDurs and then dried 8t 350~C. for about fo~r hours.
The chemical analysis of ~he silica gel for iron, aluminum, 30dium and calcium impurities following the above-described treatment was as follows: -Iro~ as Fe203 O.01Z + 0.004%
Alllmi~lUm AS A1203 0 . 01% ~ O . 004%
'Sodium as Na20 0.017. + O.Q04%
Calc~um as CaO OnO2% + 0~01%
'rable I whlch follows provides the rate of .. . ... . .. ... . . ., ,.,, . .. . . . ..... .. .. .... . .. _,. .. .. ....... .__ ., . ~

.. ..

3 ~ ~ 2 ~
12,641 product formatlon and the reaction carbon efflclency to product3 for the above-described rhDdi-~-lron cAtalyst.
As note~ from Table 1, the ~mine product mixture was comprised solely of smines ~n accordance with the invention.

.... ,. . . .. , . . .. .. . . .. . .. , ,.". ~ ~ .

.. .. ~ . -'3L133~
12,641 ~ o ? ~ ~ o . ' 1 u ~ o 8 o ~ ~ ~ o oO~
~ e . ~ o H _ ~C O
a ~ ~n E
a o o s v~ b c I v ~ O , ~ ~ O O O O O r~ o o u e ~c u G G :
~5 E IJ ~, ~ ,_ C~,~ ~ ~ , ~
¢~
e 3 ~ 0 ~ ~ o 0 ~
o~ ~ ~ c C~ rl ~ G ~ e~
~' ~ ~c~1lU ~ sp~ C1~ 1 ~6cr~ o~oSS~p~
~11 CV ~1 U I I I
Z ~Z; Z ~ I

~ ..

16.

Claims

12,641 WHAT IS CLAIMED IS:
l; A heterogeneous process for producing mono-amines in which the nitrogen is bonded only to methyl, ethyl, hydrogen or combinations thereof which comprises contacting a mixture of carbon monoxide, hydrogen and a nitrogen-containing compound selected from the group consisting of ammonia, nitric oxide and mixtures there-of with a heterogeneous solid catalyst comprising rhodium and iron at reaction conditions which comprise a temperature of from about 200° to about 450°C. a pressure of from about 15 to about 10,000 psig and a mole ratio of hydrogen to carbon monoxide of from about 20:1 to about

1:20 such that the said monoamines are formed in a collective amount of at least 50 weight percent of the total amine products of the reaction.

2. The process of claim 1 wherein said re-action conditions include a temperature within the range of about 250°-350°C., a pressure within the range of about 300-5,000 psig and a mole ratio of hydrogen to carbon monoxide within the range of about 5:1 to 1:5.

3. The process of claim 1 wherein the con-version of CO is less than about one-fourth on a single pass basis.

4. The process of claim 1 wherein the space velocity of the mixture of hydrogen, carbon monoxide and nitrogen-containing compound is in excess of about 102 GHSV.

5. The process of claim 4 wherein said space velocity is within the range of about 103 to 106 GHSV.

17.